[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7AM33JCM5TCWH3MBY37EE44":3,"skill-same:01M7AM33JCM5TCWH3MBY37EE44":48},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":25,"commitSha":26,"skillDirectory":27,"forksCount":28,"repositoryHtmlUrl":29,"repositoryHomepageUrl":6,"ownerHtmlUrl":30,"ownerBlogUrl":31,"ownerDescription":32,"updatedAt":33,"install":34,"resources":47},"01M7AM33JCM5TCWH3MBY37EE44","iterative-retrieval",null,"逐步优化上下文检索以解决子代理上下文问题的模式","Cyrilthetecchieee\u002Fcyril-code-","Cyrilthetecchieee","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F219787998?v=4","01M7AF0MMMKRAA0H61XMJPV92H","Julian Cyril D ","01M7AF0MQ10AVG6QJQ1C2BC0F5",2,"MIT","https:\u002F\u002Fgithub.com\u002FCyrilthetecchieee\u002Fcyril-code-\u002Fblob\u002Fmain\u002Fsrc\u002Fcyril_code\u002Fassets\u002F.skills_repo\u002Fdocs\u002Fzh-CN\u002Fskills\u002Fiterative-retrieval\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: iterative-retrieval\ndescription: 逐步优化上下文检索以解决子代理上下文问题的模式\norigin: ECC\n---\n\n# 迭代检索模式\n\n解决多智能体工作流中的“上下文问题”，即子智能体在开始工作前不知道需要哪些上下文。\n\n## 何时激活\n\n* 当需要生成需要代码库上下文但无法预先预测的子代理时\n* 构建需要逐步完善上下文的多代理工作流时\n* 在代理任务中遇到\"上下文过大\"或\"缺少上下文\"的失败时\n* 为代码探索设计类似 RAG 的检索管道时\n* 在代理编排中优化令牌使用时\n\n## 问题\n\n子智能体被生成时上下文有限。它们不知道：\n\n* 哪些文件包含相关代码\n* 代码库中存在哪些模式\n* 项目使用什么术语\n\n标准方法会失败：\n\n* **发送所有内容**：超出上下文限制\n* **不发送任何内容**：智能体缺乏关键信息\n* **猜测所需内容**：经常出错\n\n## 解决方案：迭代检索\n\n一个逐步优化上下文的 4 阶段循环：\n\n```\n┌─────────────────────────────────────────────┐\n│                                             │\n│   ┌──────────┐      ┌──────────┐            │\n│   │  调度    │─────│  评估    │            │\n│   └──────────┘      └──────────┘            │\n│        ▲                  │                 │\n│        │                  ▼                 │\n│   ┌──────────┐      ┌──────────┐            │\n│   │  循环    │─────│  优化    │            │\n│   └──────────┘      └──────────┘            │\n│                                             │\n│        最多3次循环，然后继续                 │\n└─────────────────────────────────────────────┘\n```\n\n### 阶段 1：调度\n\n初始的广泛查询以收集候选文件：\n\n```javascript\n\u002F\u002F Start with high-level intent\nconst initialQuery = {\n  patterns: ['src\u002F**\u002F*.ts', 'lib\u002F**\u002F*.ts'],\n  keywords: ['authentication', 'user', 'session'],\n  excludes: ['*.test.ts', '*.spec.ts']\n};\n\n\u002F\u002F Dispatch to retrieval agent\nconst candidates = await retrieveFiles(initialQuery);\n```\n\n### 阶段 2：评估\n\n评估检索到的内容的相关性：\n\n```javascript\nfunction evaluateRelevance(files, task) {\n  return files.map(file => ({\n    path: file.path,\n    relevance: scoreRelevance(file.content, task),\n    reason: explainRelevance(file.content, task),\n    missingContext: identifyGaps(file.content, task)\n  }));\n}\n```\n\n评分标准：\n\n* **高 (0.8-1.0)**：直接实现目标功能\n* **中 (0.5-0.7)**：包含相关模式或类型\n* **低 (0.2-0.4)**：略微相关\n* **无 (0-0.2)**：不相关，排除\n\n### 阶段 3：优化\n\n根据评估结果更新搜索条件：\n\n```javascript\nfunction refineQuery(evaluation, previousQuery) {\n  return {\n    \u002F\u002F Add new patterns discovered in high-relevance files\n    patterns: [...previousQuery.patterns, ...extractPatterns(evaluation)],\n\n    \u002F\u002F Add terminology found in codebase\n    keywords: [...previousQuery.keywords, ...extractKeywords(evaluation)],\n\n    \u002F\u002F Exclude confirmed irrelevant paths\n    excludes: [...previousQuery.excludes, ...evaluation\n      .filter(e => e.relevance \u003C 0.2)\n      .map(e => e.path)\n    ],\n\n    \u002F\u002F Target specific gaps\n    focusAreas: evaluation\n      .flatMap(e => e.missingContext)\n      .filter(unique)\n  };\n}\n```\n\n### 阶段 4：循环\n\n使用优化后的条件重复（最多 3 个周期）：\n\n```javascript\nasync function iterativeRetrieve(task, maxCycles = 3) {\n  let query = createInitialQuery(task);\n  let bestContext = [];\n\n  for (let cycle = 0; cycle \u003C maxCycles; cycle++) {\n    const candidates = await retrieveFiles(query);\n    const evaluation = evaluateRelevance(candidates, task);\n\n    \u002F\u002F Check if we have sufficient context\n    const highRelevance = evaluation.filter(e => e.relevance >= 0.7);\n    if (highRelevance.length >= 3 && !hasCriticalGaps(evaluation)) {\n      return highRelevance;\n    }\n\n    \u002F\u002F Refine and continue\n    query = refineQuery(evaluation, query);\n    bestContext = mergeContext(bestContext, highRelevance);\n  }\n\n  return bestContext;\n}\n```\n\n## 实际示例\n\n### 示例 1：错误修复上下文\n\n```\n任务：\"修复身份验证令牌过期错误\"\n\n循环 1:\n  分发：在 src\u002F** 中搜索 \"token\"、\"auth\"、\"expiry\"\n  评估：找到 auth.ts (0.9)、tokens.ts (0.8)、user.ts (0.3)\n  优化：添加 \"refresh\"、\"jwt\" 关键词；排除 user.ts\n\n循环 2:\n  分发：搜索优化后的关键词\n  评估：找到 session-manager.ts (0.95)、jwt-utils.ts (0.85)\n  优化：上下文已充分（2 个高相关文件）\n\n结果：auth.ts、tokens.ts、session-manager.ts、jwt-utils.ts\n```\n\n### 示例 2：功能实现\n\n```\n任务：\"为API端点添加速率限制\"\n\n周期 1：\n  分发：在 routes\u002F** 中搜索 \"rate\"、\"limit\"、\"api\"\n  评估：无匹配项 - 代码库使用 \"throttle\" 术语\n  优化：添加 \"throttle\"、\"middleware\" 关键词\n\n周期 2：\n  分发：搜索优化后的术语\n  评估：找到 throttle.ts (0.9)、middleware\u002Findex.ts (0.7)\n  优化：需要路由模式\n\n周期 3：\n  分发：搜索 \"router\"、\"express\" 模式\n  评估：找到 router-setup.ts (0.8)\n  优化：上下文已足够\n\n结果：throttle.ts、middleware\u002Findex.ts、router-setup.ts\n```\n\n## 与智能体集成\n\n在智能体提示中使用：\n\n```markdown\n在为该任务检索上下文时：\n1. 从广泛的关键词搜索开始\n2. 评估每个文件的相关性（0-1 分制）\n3. 识别仍缺失哪些上下文\n4. 优化搜索条件并重复（最多 3 个循环）\n5. 返回相关性 >= 0.7 的文件\n\n```\n\n## 最佳实践\n\n1. **先宽泛，后逐步细化** - 不要过度指定初始查询\n2. **学习代码库术语** - 第一轮循环通常能揭示命名约定\n3. **跟踪缺失内容** - 明确识别差距以驱动优化\n4. **在“足够好”时停止** - 3 个高相关性文件胜过 10 个中等相关性文件\n5. **自信地排除** - 低相关性文件不会变得相关\n\n## 相关\n\n* [长篇指南](https:\u002F\u002Fx.com\u002Faffaanmustafa\u002Fstatus\u002F2014040193557471352) - 子代理编排章节\n* `continuous-learning` 技能 - 适用于随时间改进的模式\n* 与 ECC 捆绑的代理定义（手动安装路径：`agents\u002F`）\n","# 迭代检索模式\n\n解决多智能体工作流中的“上下文问题”，即子智能体在开始工作前不知道需要哪些上下文。\n\n## 何时激活\n\n* 当需要生成需要代码库上下文但无法预先预测的子代理时\n* 构建需要逐步完善上下文的多代理工作流时\n* 在代理任务中遇到\"上下文过大\"或\"缺少上下文\"的失败时\n* 为代码探索设计类似 RAG 的检索管道时\n* 在代理编排中优化令牌使用时\n\n## 问题\n\n子智能体被生成时上下文有限。它们不知道：\n\n* 哪些文件包含相关代码\n* 代码库中存在哪些模式\n* 项目使用什么术语\n\n标准方法会失败：\n\n* **发送所有内容**：超出上下文限制\n* **不发送任何内容**：智能体缺乏关键信息\n* **猜测所需内容**：经常出错\n\n## 解决方案：迭代检索\n\n一个逐步优化上下文的 4 阶段循环：\n\n```\n┌─────────────────────────────────────────────┐\n│                                             │\n│   ┌──────────┐      ┌──────────┐            │\n│   │  调度    │─────│  评估    │            │\n│   └──────────┘      └──────────┘            │\n│        ▲                  │                 │\n│        │                  ▼                 │\n│   ┌──────────┐      ┌──────────┐            │\n│   │  循环    │─────│  优化    │            │\n│   └──────────┘      └──────────┘            │\n│                                             │\n│        最多3次循环，然后继续                 │\n└─────────────────────────────────────────────┘\n```\n\n### 阶段 1：调度\n\n初始的广泛查询以收集候选文件：\n\n```javascript\n\u002F\u002F Start with high-level intent\nconst initialQuery = {\n  patterns: ['src\u002F**\u002F*.ts', 'lib\u002F**\u002F*.ts'],\n  keywords: ['authentication', 'user', 'session'],\n  excludes: ['*.test.ts', '*.spec.ts']\n};\n\n\u002F\u002F Dispatch to retrieval agent\nconst candidates = await retrieveFiles(initialQuery);\n```\n\n### 阶段 2：评估\n\n评估检索到的内容的相关性：\n\n```javascript\nfunction evaluateRelevance(files, task) {\n  return files.map(file => ({\n    path: file.path,\n    relevance: scoreRelevance(file.content, task),\n    reason: explainRelevance(file.content, task),\n    missingContext: identifyGaps(file.content, task)\n  }));\n}\n```\n\n评分标准：\n\n* **高 (0.8-1.0)**：直接实现目标功能\n* **中 (0.5-0.7)**：包含相关模式或类型\n* **低 (0.2-0.4)**：略微相关\n* **无 (0-0.2)**：不相关，排除\n\n### 阶段 3：优化\n\n根据评估结果更新搜索条件：\n\n```javascript\nfunction refineQuery(evaluation, previousQuery) {\n  return {\n    \u002F\u002F Add new patterns discovered in high-relevance files\n    patterns: [...previousQuery.patterns, ...extractPatterns(evaluation)],\n\n    \u002F\u002F Add terminology found in codebase\n    keywords: [...previousQuery.keywords, ...extractKeywords(evaluation)],\n\n    \u002F\u002F Exclude confirmed irrelevant paths\n    excludes: [...previousQuery.excludes, ...evaluation\n      .filter(e => e.relevance \u003C 0.2)\n      .map(e => e.path)\n    ],\n\n    \u002F\u002F Target specific gaps\n    focusAreas: evaluation\n      .flatMap(e => e.missingContext)\n      .filter(unique)\n  };\n}\n```\n\n### 阶段 4：循环\n\n使用优化后的条件重复（最多 3 个周期）：\n\n```javascript\nasync function iterativeRetrieve(task, maxCycles = 3) {\n  let query = createInitialQuery(task);\n  let bestContext = [];\n\n  for (let cycle = 0; cycle \u003C maxCycles; cycle++) {\n    const candidates = await retrieveFiles(query);\n    const evaluation = evaluateRelevance(candidates, task);\n\n    \u002F\u002F Check if we have sufficient context\n    const highRelevance = evaluation.filter(e => e.relevance >= 0.7);\n    if (highRelevance.length >= 3 && !hasCriticalGaps(evaluation)) {\n      return highRelevance;\n    }\n\n    \u002F\u002F Refine and continue\n    query = refineQuery(evaluation, query);\n    bestContext = mergeContext(bestContext, highRelevance);\n  }\n\n  return bestContext;\n}\n```\n\n## 实际示例\n\n### 示例 1：错误修复上下文\n\n```\n任务：\"修复身份验证令牌过期错误\"\n\n循环 1:\n  分发：在 src\u002F** 中搜索 \"token\"、\"auth\"、\"expiry\"\n  评估：找到 auth.ts (0.9)、tokens.ts (0.8)、user.ts (0.3)\n  优化：添加 \"refresh\"、\"jwt\" 关键词；排除 user.ts\n\n循环 2:\n  分发：搜索优化后的关键词\n  评估：找到 session-manager.ts (0.95)、jwt-utils.ts (0.85)\n  优化：上下文已充分（2 个高相关文件）\n\n结果：auth.ts、tokens.ts、session-manager.ts、jwt-utils.ts\n```\n\n### 示例 2：功能实现\n\n```\n任务：\"为API端点添加速率限制\"\n\n周期 1：\n  分发：在 routes\u002F** 中搜索 \"rate\"、\"limit\"、\"api\"\n  评估：无匹配项 - 代码库使用 \"throttle\" 术语\n  优化：添加 \"throttle\"、\"middleware\" 关键词\n\n周期 2：\n  分发：搜索优化后的术语\n  评估：找到 throttle.ts (0.9)、middleware\u002Findex.ts (0.7)\n  优化：需要路由模式\n\n周期 3：\n  分发：搜索 \"router\"、\"express\" 模式\n  评估：找到 router-setup.ts (0.8)\n  优化：上下文已足够\n\n结果：throttle.ts、middleware\u002Findex.ts、router-setup.ts\n```\n\n## 与智能体集成\n\n在智能体提示中使用：\n\n```markdown\n在为该任务检索上下文时：\n1. 从广泛的关键词搜索开始\n2. 评估每个文件的相关性（0-1 分制）\n3. 识别仍缺失哪些上下文\n4. 优化搜索条件并重复（最多 3 个循环）\n5. 返回相关性 >= 0.7 的文件\n\n```\n\n## 最佳实践\n\n1. **先宽泛，后逐步细化** - 不要过度指定初始查询\n2. **学习代码库术语** - 第一轮循环通常能揭示命名约定\n3. **跟踪缺失内容** - 明确识别差距以驱动优化\n4. **在“足够好”时停止** - 3 个高相关性文件胜过 10 个中等相关性文件\n5. **自信地排除** - 低相关性文件不会变得相关\n\n## 相关\n\n* [长篇指南](https:\u002F\u002Fx.com\u002Faffaanmustafa\u002Fstatus\u002F2014040193557471352) - 子代理编排章节\n* `continuous-learning` 技能 - 适用于随时间改进的模式\n* 与 ECC 捆绑的代理定义（手动安装路径：`agents\u002F`）\n",{"name":5,"origin":24,"description":7},"ECC","main","57e7aa6b6e03bce1a7d7862ba9c2ec312f7f19b5","src\u002Fcyril_code\u002Fassets\u002F.skills_repo\u002Fdocs\u002Fzh-CN\u002Fskills\u002Fiterative-retrieval",0,"https:\u002F\u002Fgithub.com\u002FCyrilthetecchieee\u002Fcyril-code-","https:\u002F\u002Fgithub.com\u002FCyrilthetecchieee","https:\u002F\u002Fwithcyril.me\u002F","Hi, I’m Julian Cyril D, a Computer Science and Business Systems student at PSNACET.\r\nI’m passionate about full-stack development, UI\u002FUX design, and building sma","2026-11-04 23:28:50",{"source":8,"skillName":5,"commands":35,"prompt":45,"zipAvailable":46},[36,39,42],{"tool":37,"command":38},"npx","npx skills add Cyrilthetecchieee\u002Fcyril-code- --skill iterative-retrieval",{"tool":40,"command":41},"bunx","bunx skills add Cyrilthetecchieee\u002Fcyril-code- --skill iterative-retrieval",{"tool":43,"command":44},"pnpm","pnpm dlx skills add Cyrilthetecchieee\u002Fcyril-code- --skill iterative-retrieval","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FCyrilthetecchieee\u002Fcyril-code-\u002Fblob\u002Fmain\u002Fsrc\u002Fcyril_code\u002Fassets\u002F.skills_repo\u002Fdocs\u002Fzh-CN\u002Fskills\u002Fiterative-retrieval\u002FSKILL.md",true,[],[49,58,65,73,80,87],{"skillNo":50,"name":51,"nameZh":6,"description":52,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":53,"licenseFileUrl":54,"sourceHtmlUrl":55,"categories":56,"keywords":57,"publishedAt":19},"01M7AQV6CYX8RMC78420V9WZ62","webapp-testing","Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.","Apache-2.0","https:\u002F\u002Fgithub.com\u002FCyrilthetecchieee\u002Fcyril-code-\u002Fblob\u002Fmain\u002Fsrc\u002Fcyril_code\u002Fassets\u002F.skills_repo\u002Fwebapp-testing\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002FCyrilthetecchieee\u002Fcyril-code-\u002Fblob\u002Fmain\u002Fsrc\u002Fcyril_code\u002Fassets\u002F.skills_repo\u002Fwebapp-testing\u002FSKILL.md",[],[],{"skillNo":59,"name":60,"nameZh":6,"description":61,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":62,"categories":63,"keywords":64,"publishedAt":19},"01M7AQV2895N3S1MEX6K76G6G2","youtube-downloader","Download YouTube videos with customizable quality and format options. Use this skill when the user asks to download, save, or grab YouTube videos. Supports various quality settings (best, 1080p, 720p, 480p, 360p), multiple formats (mp4, webm, mkv), and audio-only downloads as MP3.","https:\u002F\u002Fgithub.com\u002FCyrilthetecchieee\u002Fcyril-code-\u002Fblob\u002Fmain\u002Fsrc\u002Fcyril_code\u002Fassets\u002F.skills_repo\u002Fvideo-downloader\u002FSKILL.md",[],[],{"skillNo":66,"name":67,"nameZh":6,"description":68,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":53,"licenseFileUrl":69,"sourceHtmlUrl":70,"categories":71,"keywords":72,"publishedAt":19},"01M7AQTRNX62DNKVZ5GXEHRBKB","theme-factory","Toolkit for styling artifacts with a theme. These artifacts can be slides, docs, reportings, HTML landing pages, etc. There are 10 pre-set themes with colors\u002Ffonts that you can apply to any artifact that has been creating, or can generate a new theme on-the-fly.","https:\u002F\u002Fgithub.com\u002FCyrilthetecchieee\u002Fcyril-code-\u002Fblob\u002Fmain\u002Fsrc\u002Fcyril_code\u002Fassets\u002F.skills_repo\u002Ftheme-factory\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002FCyrilthetecchieee\u002Fcyril-code-\u002Fblob\u002Fmain\u002Fsrc\u002Fcyril_code\u002Fassets\u002F.skills_repo\u002Ftheme-factory\u002FSKILL.md",[],[],{"skillNo":74,"name":75,"nameZh":6,"description":76,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":77,"categories":78,"keywords":79,"publishedAt":19},"01M7AQTNNVHAAN9H6DA8T6ZVE0","template-skill","Replace with description of the skill and when Claude should use it.","https:\u002F\u002Fgithub.com\u002FCyrilthetecchieee\u002Fcyril-code-\u002Fblob\u002Fmain\u002Fsrc\u002Fcyril_code\u002Fassets\u002F.skills_repo\u002Ftemplate-skill\u002FSKILL.md",[],[],{"skillNo":81,"name":82,"nameZh":6,"description":83,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":84,"categories":85,"keywords":86,"publishedAt":19},"01M7AQTJ3MQ6AKX152MEX2MYCC","tailored-resume-generator","Analyzes job descriptions and generates tailored resumes that highlight relevant experience, skills, and achievements to maximize interview chances","https:\u002F\u002Fgithub.com\u002FCyrilthetecchieee\u002Fcyril-code-\u002Fblob\u002Fmain\u002Fsrc\u002Fcyril_code\u002Fassets\u002F.skills_repo\u002Ftailored-resume-generator\u002FSKILL.md",[],[],{"skillNo":88,"name":89,"nameZh":6,"description":90,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":53,"licenseFileUrl":91,"sourceHtmlUrl":92,"categories":93,"keywords":94,"publishedAt":19},"01M7AQTD7D5MYET2YWG7HG9WM3","slack-gif-creator","Toolkit for creating animated GIFs optimized for Slack, with validators for size constraints and composable animation primitives. This skill applies when users request animated GIFs or emoji animations for Slack from descriptions like \"make me a GIF for Slack of X doing Y\".","https:\u002F\u002Fgithub.com\u002FCyrilthetecchieee\u002Fcyril-code-\u002Fblob\u002Fmain\u002Fsrc\u002Fcyril_code\u002Fassets\u002F.skills_repo\u002Fslack-gif-creator\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002FCyrilthetecchieee\u002Fcyril-code-\u002Fblob\u002Fmain\u002Fsrc\u002Fcyril_code\u002Fassets\u002F.skills_repo\u002Fslack-gif-creator\u002FSKILL.md",[],[]]